Synthesis and evaluation of novel 8-oxo-8H-cyclopenta[a]acenaphthylene-7- carbonitriles as long-wavelength fluorescent markers for hypoxic cells in solid tumor

Article abstract of DOI:10.1016/j.bmcl.2005.12.031

Novel bioreductive and long-wavelength fluorescent markers for hypoxic cells in solid tumor, 9-isocyano-8H-acenaphtho[l,2-b]pyrrol-8-one with the side chain of 2-nitroimidazole, were designed, synthesized, and evaluated in V79 379A Chinese hamster cells in vitro. Compounds A₂ and A₄ showed good hypoxic-oxic fluorescence differential in vitro (V79 cells) by using fluorescence scan ascent.

Full text of DOI:10.1016/j.bmcl.2005.12.031

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1562–1566
Synthesis and evaluation of novel
8-oxo-8H-cyclopenta[a]acenaphthylene-7-carbonitriles as long-
wavelength fluorescent markers for hypoxic cells in solid tumor
a,c,
b
a
a
Yan Liu,a,c Yufang Xu,a,c, Xuhong Qian,
Yi Xiao, Jianwen Liu, Liyun Shen,
*
*
Junhui Lid and Yuanxing Zhangd
aSchool of Pharmacy, East China University of Science and Technology, Shanghai 200237, China
bState Key Lab. Fine Chemicals, Dalian University of Technology, Dalian 116012
cShanghai Key Lab. Chem. Biology, Shanghai 200237, China
dState Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, China
Received 31 August 2005; revised 27 November 2005; accepted 9 December 2005
Available online 5 January 2006
Abstract—Novel bioreductive and long-wavelength fluorescent markers for hypoxic cells in solid tumor, 9-isocyano-8H-acenaph-
tho[l,2-b]pyrrol-8-one with the side chain of 2-nitroimidazole, were designed, synthesized, and evaluated in V79 379A Chinese
hamster cells in vitro. Compounds A2 and A4 showed good hypoxic–oxic fluorescence differential in vitro (V79 cells) by using
fluorescence scan ascent.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Most of the solid tumor cells are hypoxic. There is now
clear evidence that the oxygenation status of cells in
tumors can influence the response of these tumors to
therapy. Hypoxic cells are more resistant to radiation
than cells under well-oxygenated conditions.1 Tumor
oxygenation status has also been shown to be an
important factor in the response of tumors to certain
chemotherapeutic agents, cytokines, hyperthermia, and
photodynamic therapy. Therefore, if one could
accurately measure the oxygenation status of individual
tumors, one should be able to better predict treatment
outcome and select appropriate therapies to improve it.2
nitroaromatic compounds. The nitro group quenches
the fluorescence of the aromatic ring system, but on
bioreduction of the nitro group in hypoxic cells the
compound becomes more fluorescent. Numerous nitroa-
romatic structures have been evaluated in model exper-
iments in vitro (Scheme 1).4 Although some promising
N
O
O
O
NO2
NO2
N
(CH2)n
N
Up to now, a number of methods have been proposed
for measuring tumor oxygenation status, such as oxygen
microelectrodes, histomorphometric analysis, DNA
strand breaks, etc., however most of them invasive and
not readily available to most investigators.3 A simpler
and easier especially non-invasive method for identify-
ing hypoxic cells is suggested to be the use of fluorescent
R
N
n
O
O
N
NO2
R
O
O
NHR
Me
Me
N
N
Keywords: Fluorescent markers; Synthesis; Evaluation; Hypoxic cells;
Solid tumor.
Me
n
NO2
*
Corresponding authors. Tel.: +86 (0)21 64253589; fax: +86 (0)21
64252603; e-mail addresses: yfxu@ecust.edu.cn; xhqian@ecust.
edu.cn
Scheme 1. Structures of some reported fluorescent markers for
hypoxic cells.
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.12.031

Y. Liu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1562–1566
1563
O
results have been obtained by using flow cytometry,
CN
which is expensive and not easily available to common
researchers, the maximal absorption wavelengths of
most compounds are less than 500 nm, the low sensitiv-
ity of the eye to the bluish fluorescence from these com-
pounds making them less suitable for investigations
involving microscopy.5 In fact, the background emission
(generally in short-wavelength range) from biological
systems might be the interference in the accurate detec-
tion. Therefore, the compounds with long absorption or
fluorescence wavelength are thought to be more suitable
as fluorescent markers for hypoxic cells, and the relevant
approach by using fluorescence scan ascent, which is
cheaper and easily available, is also important and
significant for measuring tumor oxygenation status.
NC
NC
O
O
O
b
a
2
1
O
O
O
NC
NC
NC
e
c
d
HN
8-Oxo-8H-cyclopenta[a]acenaphthylene-7-carbonitrile is
a kind of novel fluorescent chromophore with long
absorption and fluorescence wavelength, whose excita-
tion and emission wavelengths can reach 530 nm and
590 nm, respectively.6 Another characteristic is its lower
DNA intercalating properties, which could avoid prob-
lems in vivo, as a compound with the high DNA affinity
could usually make the probe become localized to blood
vessels or the nuclear region of well-oxygenated cells in
tumors and fail to reach those hypoxic cells. Therefore,
in this study, 8-oxo-8H-cyclopenta[a]acenaphthylene-7-
carbonitrile bound with a side chain of 2-nitroimidazole
(A2 and A4) or 3-nitro-l,2,4-triazole (A1 and A3) as
bioreductive moiety for hypoxic cells in solid tumor
was designed, synthesized, and evaluated (Scheme 2).
N
HN
HN
N
Br
OH
5
4
3
NO2
Scheme 3. Syntheses of target compounds. Reagents and conditions:
(a) malononitrile, CH2Cl2, silica gel, rt, 95% yield; (b) K2CO3,
CH3CN, reflux, 1 h, 90% yield; (c) NH2CH2CH2OH, CH2C12, rt,
1 h, 75% yield; (d) Br2/PCl3, ethyl acetate, reflux, 5 h, 85% yield; (e) 2-
nitroimidazole or 3-nitro-1,2,4-triazole, CH3ONa/DMF, reflux, 8 h,
35% yield.
a solution of (4) in DMF was added to the mixture of
sodium methoxide and 2-nitroimidazole, and heated at
150 ꢁC for 8 h to give target compounds (A1–A4). All
the structures were confirmed by IR, 1H NMR, and
HR-MS.7
These compounds were synthesized from acenaphthyl-
ene- 1,2-dione shown in Scheme 3 (with compound A2
as an example). Acenaphthylene-l,2-dione reacted with
malononitrile at room temperature in CH2C12 and silica
gel to give red solid (1), which was refluxed for 1 h in
CH3CN in the presence of K2CO3 as a catalyst to afford
umber solid of 8-oxo-8H-cyclopenta[a]acenaphthylene-
7-carbonitrile (2). Then this solid reacted with
NH2CH2CH2OH in CH2C12 at room temperature for
1 h to give 3-(2-hydroxyethylamino)-8-oxo-8H-cyclo-
penta[a]acenaphthylene-7-carbonitrile (3). And (3)
reacted with bromine in ethyl acetate with PCl3 as a cat-
alyst to provide 3-(2-bromoethylamino)-8-oxo-8H-
cyclopenta[a]acenaphthylene-7-carbonitrile (4). Finally,
The structures of novel compounds are shown in
Scheme 1. It was found from Table 1 that the maximal
absorption and emission wavelengths of compounds
A1–A4 were about 510 and 592 nm, respectively, and
their fluorescences were still strong, which mean that
nitro group could not quench the fluorescence of the
aromatic ring system completely.
V79 379A Chinese hamster cells were used for evalua-
tion of the compounds, which were maintained as expo-
nentially growing suspension cultures in Eagle’s minimal
essential medium with Earle’s salts, modified for suspen-
sion cultures with 7.5% fetal calf serum. The compounds
were added to cell suspensions to give the appropriate
concentration at 10ꢀ4 M. Then the suspension was incu-
bated in special gases (air + 5% CO2, nitrogen + 5%
CO2) at 37 ꢁC.4 Samples from hypoxic and oxic cell sus-
pensions incubated with the solution of compounds for
different times were initially evaluated by fluorescence
a,b
Table 1. Spectra data of compounds A1–A4
Compound
UV kmax (lge)
FL kmax (/)
A1
A2
A3
A4
515 (4.48), 566 (4.93)
510 (4.54), 560 (4.91)
517 (4.49), 567 (4.92)
511 (4.47), 562 (4.92)
591 (0.16)
591 (0.13)
592 (0.19)
592 (0.15)
a In acetone.
b With quinine sulfate in sulfuric acid as quantum yield standard
(/ = 0.55).
Scheme 2. Novel fluorescent markers for hypoxic cells.

1564
Y. Liu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1562–1566
microscopy using 510 nm excitation wavelengths. Then
a quantitative study of the time courses of accumulation
of fluorescent metabolites in the same condition was car-
ried out with fluorescence scan ascent. The results are
shown in Figure 1.
imidazole moiety to the naphthalimide nucleus of A4.8 It
was reported that the PET efficiency was related to sev-
eral aspects:9,10 (a) It was directly related to the presence
of the electron donors. That is, the electron-deficient
moiety of 2-nitroimidazole would become electron-rich
moiety of 2-aminoimidazole after bioreduction, which
was easy to proceed with PET to quench fluorescence
somewhat and balance the fluorescence enhancement.
(b) Longer spacer can also weaken the quenching effi-
ciency by preventing effective approach of electron from
the donor to the receptor. In this paper, the side-chain
length of A2 was shorter than that of A4, PET efficiency
of the reduction product of A2 was higher than that of
A4, and fluorescence quenching efficiency of A2’s reduc-
tive product should be also higher than that of A4, so
that hypoxic–oxic fluorescence differential of A2 was
lower than that of A4.
From Figure 1, obvious differential of the fluorescence
can be seen between the cells incubated under hypoxic
and oxic conditions. It was found by using fluorescence
scan ascent that the fluorescence of cells incubated under
oxic condition had little change all the time, but the case
was different with hypoxic cells. The compounds A2 and
A4 (with 2-nitroimidazole as reductive moiety) in hypox-
ic cells were reduced by nitroreductase enzymes in two-
electron steps in an oxygen-insensitive process, which
resulted in significant fluorescence enhancement of cells
incubated during three hours. After 3 h, the fluorescence
in cells incubated reached the maximal value and then
kept it. Of course, the hypoxic–oxic differential reached
the largest at the same time.
The study of the time courses of accumulation of fluo-
rescent metabolites in V79 cells incubated with A1 and
A3, was also carried out by using fluorescence scan as-
cent, the hypoxic–oxic fluorescence differential incubat-
ed with A1 and A3 being 4 times and 6 times,
respectively. The difference between A1 and A3 was
caused by the same two factors as the case of A2 and A4.
The largest hypoxic–oxic fluorescence differential of A4
in cells (15 times) was higher than that of A2 (11 times).
The difference might have been caused by two factors:
(1) The introduction of an ether moiety as a hydrophilic
group into the side chain of A4 promoted its water sol-
ubility, thus improving the transport of compound in vi-
vo to tumor cells. (2) Weak fluorescence quenching
caused by a slight intramolecular photoinduced electron
transfer (PET) from the reduction products of the nitro-
The difference in fluorescence behavior between two
kinds of compounds, A2, A4 and A1 A3, might be due
to that of their bioreductive moieties in the side chains.
2-Nitroimidazole was a strong electron-deficient moiety
by comparison with 3-nitro-l,2,4-triazole,8 but the elec-
tron-donating ability for bioreductive product (corre-
sponding to the PET efficiency for fluorescence
quenching to balance the fluorescence enhancement dur-
ing bioreduction) of the latter was higher than that of
the former, which finally make A1 and A3 have small
hypoxic–oxic fluorescence differential in cells. From
the redox point of view, the oxidation potential of the
electron donor species is one of the most important fac-
tors affecting the PET efficiency.11 The strong electron-
rich reductive product of 3-nitro-l,2,4-triazole was easier
to lose electron for oxidation than the bioreductive
product of 2-nitroimidazole did, which resulted in a
higher PET efficiency of A1 and A3.
a
14
12
10
8
hypoxic
oxic
6
4
2
0
0
1
2
3
4
5
6
By fluorescence microscopy, we also got the fluorescence
microphotographs of V79 cells incubated with 10ꢀ4 M
of A2 and A4, respectively (Figs. 2 and 3).
Time/hr
b
16
14
12
10
8
hypoxic
oxic
It was found by using fluorescence microscopy that the
cells incubated with compounds (A1–A4) showed red
fluorescence under irradiation of light at 510 nm,
although hypoxic–oxic fluorescence differential in the
images was not obvious enough (Figs. 2 and 3). Howev-
er, more importantly, obvious differential fluorescence
between hypoxic and oxic cells was observed and
recorded by using fluorescence scan ascent (Fig. 1). In
fact, a number of nitroaromatic compounds have been
evaluated as fluorescent markers for hypoxic cells
in model experiments in vitro, emissions for most of
the reduced products in hypoxic cells being in short
wavelength, for example, blue, green, and yellow
fluorescence.
6
4
2
0
Time/hr
Figure 1. The time courses of accumulation of fluorescent metabolites
in V79 379A Chinese hamster cells incubated with 10ꢀ4 M compounds
at 37 ꢁC. (a) A2; (b) A4.

Y. Liu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1562–1566
1565
in V79 cells could reach 11 times and 15 times, respec-
tively, which revealed they might be promising candi-
date markers for hypoxic cells.
Acknowledgments
The National Basic Research Program of China
(2003CB114400), National Natural Science Foundation
of China, The Science and Technology Foundation of
Shanghai, and The Education Commission of Shanghai
partially supported this study.
References and notes
1. (a) Brizel, D. M.; Scully, S. P.; Harrelson, J. M.; Layfield,
L. J.; Bean, J. M.; Prosnitz, L. R.; Dewhirst, M. W.
Cancer Res. 1996, 56, 941; (b) Phillips, T. L.; Wasserman,
T. H.; Stetz, J.; Brady, L. W. Int. J. Radiat. Oncol. Biol.
Phys. 1982, 8, 327; (c) Oliver, P. L.; Ralph, E. D. Cancer
Res. 1983, 43, 3276; (d) Vaupel, P.; Kallinowski, F.;
Okunieff, P. Cancer Res. 1989, 49, 6449.
Figure 2. Fluorescence microphotographs of V79 cells incubated with
10ꢀ4 M of A2 at 37 ꢁC. After 3.5 h incubation, scanning was taken.
Magnification was 1000·. (a) Scanning was taken on brightfield, cells
under hypoxic condition (incubated in nitrogen and 5% CO2); (b)
excited at 510 nm, cells under hypoxic condition; (c) scanning was
taken on brightfield, cells under oxic condition (incubated in air and
5% CO2); (d) excited at 510 nm, cells under oxic condition.
2. Michael, R. H. Int. J. Radiat. Oncol. Biol. Phys. 1998, 42,
701.
3. (a) Fyles, A. W.; Miolsevic, M.; Wong, R.; Kavanagh, M.
C.; Pintilie, M.; Sun, A.; Chapman, W.; Levin, W.;
Manchul, L.; Keane, T. J.; Hill, R. P. Radiother. Oncol.
1998, 48, 149; (b) Al-Hallaq, H. A.; River, J. N.; Zamora,
M.; Oikawa, H.; Karczmar, G. S. Int. J. Radiat. Oncol.
Biol. Phys. 1998, 41, 151; (c) Karen, A. Y.; Siham, B. P.;
Rachelle, M. L.; Darrell, Q. B.; Matthew, C. F.; James, S.
B.; Gerald, E. H.; Chapman, J. D. Int. J. Radiat. Oncol.
Biol. Phys. 1995, 33, 111.
4. (a) Hodgkiss, R. J.; Begg, A. C.; Middleton, R. W.;
Parrick, J.; Stratford, M. R. L.; Wardman, P.; Wilson, G.
D. Biochem. Pharmacol. 1991, 41, 533; (b) Hodgkiss, R. J.;
Middleton, R. W.; Parrick, J.; Rami, H. K.; Wardman, P.;
Wilson, G. D. J. Med. Chem. 1992, 35, 1920; (c) Hodgkiss,
R. J.; Parrick, J.; Manuchehr, P.; Stratford, M. R. L.
J. Med. Chem. 1994, 37, 4352.
5. Stratford, M. R. L.; Clarke, E. D.; Hodgkiss, R. J.;
Middleton, R. W.; Wardman, P. Int J. Radiat. Oncol. Biol.
Phys. 1984, 10, 1353.
6. Xiao, Y.; Liu, F.; Qian, X. Chem. Commun. 2005, 2, 239.
7. Compound A1: mp > 300 ꢁC. 1H NMR (SF: 500 MHz,
CDCl3-d1), d (ppm), 8.86 (d, 1H, J = 8.64 Hz), 8.75 (d,
1H, J = 7.29 Hz), 8.72 (d, 1H, J = 8.01 Hz), 8.42 (d, 1H,
J = 7.99 Hz), 7.99 (t, 1H, J1 = 7.43 Hz, J2 = 7.55 Hz,),
7.05 (s, 1H, triazole-50H), 4.47 (t, 2H, J1 = 5.41 Hz,
J2 = 5.39 Hz, –CH2–triazole), 4.01 (t, 2H, J1 = 5.46 Hz,
J2 = 5.33 Hz, –NHCH2–). HRMS (ESI): C19H11N7O3
Figure 3. Fluorescence microphotographs of V79 cells incubated with
10ꢀ4 M of A4 at 37 ꢁC After 3.5 h incubation, scanning was taken.
Magnification was 1000·. (a) Scanning was taken on brightfield, cells
under hypoxic condition (incubated in nitrogen and 5% CO2); (b)
excited at 510 nm, cells under hypoxic condition; (c) scanning was
taken on brightfield, cells under oxic condition (incubated in air and
5% CO2); (d) excited at 510 nm, cells under oxic condition.
calculated: 385.0923; found: 385.0925; IR (KBr), cmꢀ1
:
1
3310, 1672, 1595, 1570. Compound A2: mp > 300 ꢁC. H
NMR (SF: 500 MHz, CDCl3-d1), d (ppm), 9.35 (d, 1H,
J = 2.00 Hz), 9.16 (d, 1H, J = 2.05 Hz), 8.80 (d, 1H,
J = 7.29 Hz), 8.45 (d, 1H, J = 7.25 Hz), 7.97 (t, 1H,
J1 = 7.79 Hz, J2 = 7.75 Hz), 7.92 (d, 1H, J = 7.48,
imidazole- 40H), 6.78 (d, H, J = 8.43, imidazole-50H), 4.50
(t, 2H, J1 = 4.48 Hz, J2 = 5.38 Hz, –CH2–imidazole), 4.01
(t, 2H, J1 = 5.42 Hz, J2 = 5.42 Hz, –NHCH2–). HRMS
(ESI): C20H12N6O3 calculated: 384.0971; found: 384.0975;
In summary, the present work demonstrated novel long-
wavelength fluorescent markers for hypoxic cells in solid
tumor. Some of the compounds (A2 and A4) showed
obvious fluorescence differential between hypoxic and
oxic cells (V79 cells) in vitro by using fluorescence scan
ascent. When time was approaching 3 h, the hypoxic–
oxic fluorescence differential incubated with A2 and A4
IR (KBr), cmꢀ1
Compound A3: mp > 300 ꢁC 1H NMR (SF: 500 MHz,
MeOD-d4), (ppm), 8.42 (m, 2H,), 8.25 (d, 1H,
:
3455, 2212, 1625, 1585, 1570.
d
J = 8.56 Hz), 7.54 (d, 1H, J1 = 7.60 Hz, J2 = 8.19 Hz),
7.04 (s, 1H, triazole-50H), 6.74 (d, 1H, J = 8.56Hz), 4.31

1566
Y. Liu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1562–1566
(m, 4H, –NHCH2CH2O–), 3.78 (t, 2H, OCH2CH2–,
J1 = 5.71 Hz, J2 = 5.8 Hz), 3.49 (t, 2H, J1 = 5.76 Hz,
J2 = 5.72 Hz, -OCH2CH2–). HRMS (ESI): C21H15N7O4
8. (a) de Silva, A. P.; Gunaratne, H. Q. N.; Habib-Jiwan, J.-L;
McCoy, C. P.; Rice, T. E.; Soumillion, J.-P. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1728; (b) Baggott, J. E.; Pilling, M. J.
J. Chem. Soc., Faraday Trans. 1 1983, 79, 221.
calculated: 429.1185; found: 429.1188; IR (KBr), cmꢀ1
:
3320, 2215, 1638, 1580, 1492. Compound A4:
mp > 300 ꢁC. 1H NMR (SF: 500 MHz, acetone-d6), d
(ppm), 8.12 (d, 1H, J = 8.10 Hz), 8.06 (d, 1H,
J = 7.10 Hz), 7.82 (t, 1H, J1 = 7.53 Hz, J2 = 7.55 Hz),
7.64 (d, 2H, J = 7.45 Hz, imidazole-50H), 7.55 (d, 1H,
J1 = 7.30 Hz, J2 = 7.17 Hz), 7.25 (d, 1H, J = 7.32 Hz,
imidazole-40H), 4.11 (m, 6H, –NHCH2CH2OCH2CH2–),
3.59 (d, 2H, J = 4.7 Hz, –OCH2CH2–). HRMS (ESI):
C22H16N6O4 calculated: 428.1233; found: 428.1236; IR
(KBr), cmꢀ1: 3320, 1632, 1570, 1490.
9. Gong, Y.-K.; Nakashima, K. J. Phys. Chem. B 2002,
106, 803.
10. (a) Wilson, W. R. J. Phys. Chem. A 1997, 101, 9704; (b)
Mayaa, J. D.; Bollob, S.; Nun˜ez-Vergarab, L. J.; Squellab,
´ ´
J. A.; Repettoa, Y.; Morelloa, A.; Periec, J.; Chauvierec, G.
Biochem. Pharmacol. 2003, 65, 999; (c) Viode, C.; Bettache,
`
´
`
N.; Cenas, N.; Krauth-Siegel, R. L.; Chauviere, G.;
Bakalara, N.; Perie, J. Biochem. Pharmacol. 1999, 57, 549.
´ ´
11. Myles, A. J.; Branda, N. R. J. Am. Chem. Soc. 2001,
123, 177.